diy solar

diy solar

Asking advice on best panel size and configuration options for rooftop 2kW array - 48 volt system.

KenDan

New Member
Joined
Mar 5, 2020
Messages
77
Location
Central Pennsylvania
I'm about to order panels and realized that maybe it's not quite as simple as I thought.
I have an MPP Solar LV5048 split phase unit on order and am considering the following panel options:
20 - Rich Solar 100 watt Poly Panels - or
12 - Rich Solar 160 watt Poly Panels

My house has a moderately shallow roof pitch (about 20°) and I'm at 41N latitude, so my winter sun peak elevation can dip below 26°. I'd like to find a way to mount my panels on the roof at somewhat increased angle to better compromise summer and winter sun elevation angles. I'm thinking a good compromise would be 45° from vertical. That means I need to tilt my panels by an additional 25° or so. (I am fortunate that my house faces almost exactly due south and has a totally unobstructed view in that direction)
I need to research what types of mounting hardware is available to do this. Any suggestions for mounting panels on the roof at increased tilt angles?

Tilting the panels on the roof seems to introduce a few other potential problems.
1) Less stable and increased vulnerability to wind.
2) Rows must be kept further apart to prevent the lower row from shading the upper row in the winter.

Although 12 of the 160 watt panels would be less connections and less mounting points, it seems the larger panels would exasperate problems 1) and 2).
Is it better to go with the smaller 100 watts panels?

My other question has to do with how connect the panels into the LV5048 which has two PV inputs. My general options seem to be:

If I use 12 of the 160 watt panels - connect 8 of them (4s2p) into one PV input and the remaining string of 4 (4s1p) into the other PV input.

If I use 20 of the 100 watt panels - connect 12 of them (4s3p) into one PV input and the remaining 8 (4s2p) into the other PV input.
Or perhaps, if using 20 of the 100 watt panels it would be better to connect a 10 of them (5s2p) into each PV input?

Thanks!
 
Last edited:
Most racking companies have stopped making tilt legs for roofs due to the complications you listed. Just install more solar panels if you have room to make up the difference. Installing them flat will allow you to install more without the spacing. Why would you use 12V (36 cell) panels for your system? Use 60 or 72 cell panels for a fraction of the price, the MPPT inputs will drop the voltage of at least 3 in series to the 48V battery bank. I couldn't find the max voltage input, so don't know how many in series you can do.
 
Most racking companies have stopped making tilt legs for roofs due to the complications you listed. Just install more solar panels if you have room to make up the difference. Installing them flat will allow you to install more without the spacing. Why would you use 12V (36 cell) panels for your system? Use 60 or 72 cell panels for a fraction of the price, the MPPT inputs will drop the voltage of at least 3 in series to the 48V battery bank. I couldn't find the max voltage input, so don't know how many in series you can do.

Hmmm, having my panels tilted ~45 degrees above the maximum sun elevation during the winter (due to my roof pitch) is going to cause an large performance hit. I believe the output is the cosine of the incidence angle times the rated output. So, with a 45 degree incidence angle the output is only 70% - 70 watts each or about 1400 watts total (theoretical of course). The losses that come with the sun not being aligned laterally with the panel mount even faster than when the panel in not closely aligned vertically because of the geometry. I'd have to buy at least 8 more panels putting out 70 watts maximum - another $600+. I think I can fabricate aluminum angle tilt mounts for a lot less than that.

I can avoid the shading problem, but wind loading is still a concern... On the positive side - a 45 degree tilt from vertical would be a huge help with snow accumulation in winter.

The LV5048 can handle 145vdc max open circuit input., and 5 panels in series would be 106 volts (21.2 volts open circuit per panel). Going to 10 panels in series would put me way over voltage. Even 7 panels would put me slightly over the max if I bought 8 more for a total of 28. (28 panels in a 7s4p configuration)

I didn't realize that 60 or 72 cell panels were more cost effective.
EDIT: I just checked out the 355 watt poly panels (the best kW/$ ratio model) on the AltE site and after figuring in freight costs, the total kW/$ is very close to the $75 Rich Solar 100 watt panels. (Rich Solar is $750/kW, and the 355 watt altE polys are $732/kW including freight) Definitely a small savings, but the tradeoff is that a damaged panel in more expensive to replace, and they are a much larger wind sail. Tough call...
 
Last edited:
You can use SAM to play around with annual power as affected by tilt and a lot of other things. One of those is weather, I take a "double hit" going flat because you'd think summer is the optimum months for flat panels here with the sun at ~90° on the summer solstice, turns out it's also the stormiest season with a lot of afternoon clouds.

But, I live in a hurricane prone region. From this it seemed the force from hurricane force winds downwards would be about the same as a small herd of elephants on the roof if I went with the optimum angle. Even a little tilt made a huge difference. That's why I went flat.
 
Go to PVWatts and insert your zip code (if in USA, if not, find a zip code with the same latitude, like 06460.) For a 2kW system at 41 degrees latitude, flat on the 20 degree roof generates 2585kWh/ year. 45 degrees adds a little, 2649kWh, and going to 15 above latitude at 56 degrees gives less at 2554kWh/year. Unless you are going completely off-grid, I'm not sure that the 100kWh a year is worth it. You are right about the snow, you will have some losses in the winter unless you get a snow brush and clear it off. But quite frankly, at 41 latitude, you are not making much energy in the winter anyways.

When calculating max voltage in, be sure to compensate for cold temperatures when voltage rises. Not knowing where you are, I would be overly cautious and add 20% to the voltage. 21.2Voc becomes 25.4Voc, so you are limited to 5 12V panels, 3 20V panels, or 2 24V panels in series. Don't push the limit, overvoltage will release the factory installed smoke (boom). With that limitation, with 2 strings, you can do 10 12V 160W panels for 1600W, or 6 310W 20V panels for 1860W, or 4 24V 355W panels for 1420W (which is why 24V panels aren't usually used in residential). Of course you can double those numbers with parallel strings. Moral of the story is you can get your highest wattage total input using 60 cell panels without blowing up your inverter.
 
I'm at 34 degrees latitude on a 4:12 pitch. 10 200W panels with an LV5048. Sets of 5 in series then parallel in a midnight solar to PV input 1.
I use 4 12v 100ah battleborn. The LV5048 recommends at least 200ah but I have had no issues. I'll expand later going to 4kw of panels and 200ah of battery and utilize PV input 2
It sounds like the system I have planned is similar to yours. My roof pitch is roughly the same as yours, but you are about 7º farther south. I have a 48v 150ah battery system planned - it is hopefully "on the way". (16 aluminum cased 150 ah LiFePO4 cells).
I'm wondering, why did you decide to run both of your 5S strings of panels into PV input 1 and not split them between PV1 and PV2? Did you just do that to keep PV input 2 open because you knew you were going to be adding more panels soon?
 
Most racking companies have stopped making tilt legs for roofs due to the complications you listed. Just install more solar panels if you have room to make up the difference. Installing them flat will allow you to install more without the spacing. Why would you use 12V (36 cell) panels for your system? Use 60 or 72 cell panels for a fraction of the price, the MPPT inputs will drop the voltage of at least 3 in series to the 48V battery bank. I couldn't find the max voltage input, so don't know how many in series you can do.
Great job thank you for the clear explanations!
 
There are certainly a lot of variables involved, but only some of them can be controlled. I guess the online calculators are OK for rough ballpark estimates, but being an engineer I want to be able to actually calculate the variables I can get my arms around. This is why I'm focused on roof pitch and tilt angle. Power output vs incidence angle is a sine function and can be calculated by cosine(theta). Seasonal minimum and maximum sun incidence angles can be be determined by adding or subtracting 23.5° from your latitude. Increased atmospheric absorption at lower solar elevations in winter causes a loss in solar flux density, but decreased temperatures in winter yields a higher PV panel efficiency. These two values seem as if they'd partially cancel out.
The largest factor (assuming full sunlight is available) really does seem to be panel tilt relative to the sun. At first the effect is very small, but when you get up to 45° off-axis you have lost 30%. As the angle increases further, things go downhill faster. At 55° you lose an additional 13% (43% total loss), and at 65° you lose another 15% (58% total loss).
On a fixed array you lose these amounts already as the sun moves East and West of the central meridian (due south). If the panels are not tilted near the optimum elevation angle, that angle compounds with the East-West angle and causes even greater losses because the losses are not linear with respect to increasing angle due to the sine function.
I found this page which seems to have lots of relevant math and formulas:

Going back to my questions:). Has anyone used the Renogy Adjustable Solar Panel Tilt Mount Brackets on a home rooftop installation? I see that Amazon has them - but has a quantity limit of 8 per person:( These would save me a lot of work fabricating mounts from scratch if they are strong enough.
 
you maybe able to win on lot of fronts with technology but you cannot win against nature. with the tilt you are knowingly introducing a huge wind load risk.
your assumptions are all leaning towards a best case scenario but the tilt introduces a real worst case scenario and may end up with a hole in your roof. Also the higher wattage panels are generally reliable and last long time.
if you are worried about the output one option is to invest in more panels or clean the panel of snow when needed or very real option is add the extra panels on the side of the house facing south/west/east if that is possible.
 
... the online calculators are OK for rough ballpark estimates, but being an engineer I want to be able to actually calculate the variables I can get my arms around....

PVWatts is a subset of SAM and uses reasonable defaults. It's not a simple calculator (or calculation) by any means and does takes local weather conditions into account; but I can see where an engineer might want more. The value of something like SAM beyond PVWatts is it can estimate the total cost of lifetime ownership vs output and how they change based on small changes to the system. By manually entering values on the losses page you can answer questions like is it worth it to reduce powerloss by 0.1% by using a lower gauge of wire over the life of the system? Download SAM and use the Photovoltaic or PVWatts model, as an engineer you'll appreciate how much effort NREL built into it and it's level of complexity.

Keep in mind too that solar is highly variable (clouds and such), and the envelope of accuracy of any calculations will depend on how closely the weather follows normal patterns.

What Sam won't tell you is the wind loading.

Naturally when I started I did it backwards starting with pitch, tilt, azimuth of the panels to optimize power and minimize costs. It wasn't until near the end when I realized wind trumped all of that for my case and I was happy to accept the double loss mentioned above to avoid the elephantine forces at play.

.... [are the bracket legs] strong enough
You'd have to calculate the forces from the maximum wind in your area based on your planned installation's tilt/roof-angle to be able to answer that. You can find the maximum winds speed from a wind zone map for your county. Your county may also have adopted a specific version of ASCE 7 which you'll want to adhere to pass building inspections.
 
Last edited:
You'd have to calculate the forces from the maximum wind in your area based on your planned installation's tilt/roof-angle to be able to answer that. You can find the maximum winds speed from a wind zone map for your county. Your county may also have adopted a specific version of ASCE 7 which you'll want to adhere to pass building inspections.

Thanks for the link to the calculate the forces page! There is some very good information there that I can definitely use. I am rethinking the amount of tilt angle, and am also looking at other things that I do with my installation to minimize wind loading. I have a number of factors working in my favor with respect to being surrounded by 70 foot tall trees within 50 to 75 yards on the north east and west, and living in a single story home. Although I live in zone 3, I can't remember ever experiencing a wind gust higher than 60 mph on my weather station, and those are extremely rare. It seems that measures such as keeping panels as far away from the edge of the roof as possible, and keeping the high edge of the panel as low as possible will help to reduce the wind loading. This is one reason why I decided to go with smaller 100 watts panels. If mounted horizontally, the height of the higher edge is substantially reduced. With smaller panels, you can allow more gaps between them to bleed off some the the pressure differential. More panels and mounting brackets also distributes the forces across many more points.
I do recognize the importance of not underestimating the forces that wind can generate. It don't think it's an insurmountable problem - but one that may demand more careful consideration than I originally thought.
 
Back
Top